INDEX

Explanations

proper nouns associated with news articles and reports, including names of individuals and organizations

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Configuration

jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

standard

Context Size

1,024

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 bordeaux

-0.90

 fluo

-0.79

 oleo

-0.79

 cioc

-0.78

 soggior

-0.75

ouret

-0.74

 ecru

-0.73

uccio

-0.71

 vanil

-0.70

 Conc

-0.69

POSITIVE LOGITS

 explains

0.66

 commented

0.65

 explained

0.62

 oversaw

0.62

 disagreed

0.61

 kasarigan

0.61

 testified

0.60

 says

0.59

 remarked

0.59

 said

0.59

Activations Density 0.288%

proper nouns associated with news articles and reports, including names of individuals and organizations

No Comments

No Known Activations